server/unicode.c

   1 /*
   2  * Unicode routines for use inside the server
   3  *
   4  * Copyright (C) 1999 Alexandre Julliard
   5  *
   6  * This library is free software; you can redistribute it and/or
   7  * modify it under the terms of the GNU Lesser General Public
   8  * License as published by the Free Software Foundation; either
   9  * version 2.1 of the License, or (at your option) any later version.
  10  *
  11  * This library is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  14  * Lesser General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU Lesser General Public
  17  * License along with this library; if not, write to the Free Software
  18  * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
  19  */
  20
  21 #include "config.h"
  22
  23 #include <ctype.h>
  24 #include <fcntl.h>
  25 #include <stdio.h>
  26 #include <stdarg.h>
  27 #include <unistd.h>
  28 #include <limits.h>
  29 #ifdef HAVE_SYS_SYSCTL_H
  30 # include <sys/sysctl.h>
  31 #endif
  32
  33 #include "windef.h"
  34 #include "winternl.h"
  35 #include "request.h"
  36 #include "unicode.h"
  37 #include "file.h"
  38
  39 /* number of following bytes in sequence based on first byte value (for bytes above 0x7f) */
  40 static const char utf8_length[128] =
  41 {
  42     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x80-0x8f */
  43     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0x90-0x9f */
  44     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xa0-0xaf */
  45     0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, /* 0xb0-0xbf */
  46     0,0,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xc0-0xcf */
  47     1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, /* 0xd0-0xdf */
  48     2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, /* 0xe0-0xef */
  49     3,3,3,3,3,0,0,0,0,0,0,0,0,0,0,0  /* 0xf0-0xff */
  50 };
  51
  52 /* first byte mask depending on UTF-8 sequence length */
  53 static const unsigned char utf8_mask[4] = { 0x7f, 0x1f, 0x0f, 0x07 };
  54
  55 /* minimum Unicode value depending on UTF-8 sequence length */
  56 static const unsigned int utf8_minval[4] = { 0x0, 0x80, 0x800, 0x10000 };
  57
  58 static unsigned short *casemap;
  59
  60 static inline char to_hex( char ch )
  61 {
  62     if (isdigit(ch)) return ch - '0';
  63     return tolower(ch) - 'a' + 10;
  64 }
  65
  66 static inline WCHAR to_lower( WCHAR ch )
  67 {
  68     return ch + casemap[casemap[casemap[ch >> 8] + ((ch >> 4) & 0x0f)] + (ch & 0x0f)];
  69 }
  70
  71 int memicmp_strW( const WCHAR *str1, const WCHAR *str2, data_size_t len )
  72 {
  73     int ret = 0;
  74
  75     for (len /= sizeof(WCHAR); len; str1++, str2++, len--)
  76         if ((ret = to_lower(*str1) - to_lower(*str2))) break;
  77     return ret;
  78 }
  79
  80 unsigned int hash_strW( const WCHAR *str, data_size_t len, unsigned int hash_size )
  81 {
  82     unsigned int i, hash = 0;
  83
  84     for (i = 0; i < len / sizeof(WCHAR); i++) hash = hash * 65599 + to_lower( str[i] );
  85     return hash % hash_size;
  86 }
  87
  88 WCHAR *ascii_to_unicode_str( const char *str, struct unicode_str *ret )
  89 {
  90     data_size_t i, len = strlen(str);
  91     WCHAR *p;
  92
  93     ret->len = len * sizeof(WCHAR);
  94     ret->str = p = mem_alloc( ret->len );
  95     if (p) for (i = 0; i < len; i++) p[i] = (unsigned char)str[i];
  96     return p;
  97 }
  98
  99 /* parse an escaped string back into Unicode */
 100 /* return the number of chars read from the input, or -1 on output overflow */
 101 int parse_strW( WCHAR *buffer, data_size_t *len, const char *src, char endchar )
 102 {
 103     WCHAR *dest = buffer;
 104     WCHAR *end = buffer + *len / sizeof(WCHAR);
 105     const char *p = src;
 106     unsigned char ch;
 107
 108     while (*p && *p != endchar && dest < end)
 109     {
 110         if (*p == '\\')
 111         {
 112             p++;
 113             if (!*p) break;
 114             switch(*p)
 115             {
 116             case 'a': *dest++ = '\a'; p++; continue;
 117             case 'b': *dest++ = '\b'; p++; continue;
 118             case 'e': *dest++ = '\e'; p++; continue;
 119             case 'f': *dest++ = '\f'; p++; continue;
 120             case 'n': *dest++ = '\n'; p++; continue;
 121             case 'r': *dest++ = '\r'; p++; continue;
 122             case 't': *dest++ = '\t'; p++; continue;
 123             case 'v': *dest++ = '\v'; p++; continue;
 124             case 'x':  /* hex escape */
 125                 p++;
 126                 if (!isxdigit(*p)) *dest = 'x';
 127                 else
 128                 {
 129                     *dest = to_hex(*p++);
 130                     if (isxdigit(*p)) *dest = (*dest * 16) + to_hex(*p++);
 131                     if (isxdigit(*p)) *dest = (*dest * 16) + to_hex(*p++);
 132                     if (isxdigit(*p)) *dest = (*dest * 16) + to_hex(*p++);
 133                 }
 134                 dest++;
 135                 continue;
 136             case '0':
 137             case '1':
 138             case '2':
 139             case '3':
 140             case '4':
 141             case '5':
 142             case '6':
 143             case '7':  /* octal escape */
 144                 *dest = *p++ - '0';
 145                 if (*p >= '0' && *p <= '7') *dest = (*dest * 8) + (*p++ - '0');
 146                 if (*p >= '0' && *p <= '7') *dest = (*dest * 8) + (*p++ - '0');
 147                 dest++;
 148                 continue;
 149             }
 150             /* unrecognized escape: fall through to normal char handling */
 151         }
 152
 153
 154         ch = *p++;
 155         if (ch < 0x80) *dest++ = ch;
 156         else  /* parse utf8 char */
 157         {
 158             int charlen = utf8_length[ch-0x80];
 159             unsigned int res = ch & utf8_mask[charlen];
 160
 161             switch(charlen)
 162             {
 163             case 3:
 164                 if ((ch = *p ^ 0x80) >= 0x40) break;
 165                 res = (res << 6) | ch;
 166                 p++;
 167             case 2:
 168                 if ((ch = *p ^ 0x80) >= 0x40) break;
 169                 res = (res << 6) | ch;
 170                 p++;
 171             case 1:
 172                 if ((ch = *p ^ 0x80) >= 0x40) break;
 173                 res = (res << 6) | ch;
 174                 p++;
 175                 if (res < utf8_minval[charlen]) break;
 176                 if (res > 0x10ffff) break;
 177                 if (res <= 0xffff) *dest++ = res;
 178                 else /* we need surrogates */
 179                 {
 180                     res -= 0x10000;
 181                     *dest++ = 0xd800 | (res >> 10);
 182                     if (dest < end) *dest++ = 0xdc00 | (res & 0x3ff);
 183                 }
 184                 continue;
 185             }
 186             /* ignore invalid char */
 187         }
 188     }
 189     if (dest >= end) return -1;  /* overflow */
 190     *dest++ = 0;
 191     if (!*p) return -1;  /* delimiter not found */
 192     *len = (dest - buffer) * sizeof(WCHAR);
 193     return p + 1 - src;
 194 }
 195
 196 /* dump a Unicode string with proper escaping */
 197 int dump_strW( const WCHAR *str, data_size_t len, FILE *f, const char escape[2] )
 198 {
 199     static const char escapes[32] = ".......abtnvfr.............e....";
 200     char buffer[256];
 201     char *pos = buffer;
 202     int count = 0;
 203
 204     for (len /= sizeof(WCHAR); len; str++, len--)
 205     {
 206         if (pos > buffer + sizeof(buffer) - 8)
 207         {
 208             fwrite( buffer, pos - buffer, 1, f );
 209             count += pos - buffer;
 210             pos = buffer;
 211         }
 212         if (*str > 127)  /* hex escape */
 213         {
 214             if (len > 1 && str[1] < 128 && isxdigit((char)str[1]))
 215                 pos += sprintf( pos, "\\x%04x", *str );
 216             else
 217                 pos += sprintf( pos, "\\x%x", *str );
 218             continue;
 219         }
 220         if (*str < 32)  /* octal or C escape */
 221         {
 222             if (!*str && len == 1) continue;  /* do not output terminating NULL */
 223             if (escapes[*str] != '.')
 224                 pos += sprintf( pos, "\\%c", escapes[*str] );
 225             else if (len > 1 && str[1] >= '0' && str[1] <= '7')
 226                 pos += sprintf( pos, "\\%03o", *str );
 227             else
 228                 pos += sprintf( pos, "\\%o", *str );
 229             continue;
 230         }
 231         if (*str == '\\' || *str == escape[0] || *str == escape[1]) *pos++ = '\\';
 232         *pos++ = *str;
 233     }
 234     fwrite( buffer, pos - buffer, 1, f );
 235     count += pos - buffer;
 236     return count;
 237 }
 238
 239 static char *get_nls_dir(void)
 240 {
 241     char *p, *dir, *ret;
 242     const char *nlsdir = BIN_TO_NLSDIR;
 243
 244 #if defined(__linux__) || defined(__FreeBSD_kernel__) || defined(__NetBSD__)
 245     dir = realpath( "/proc/self/exe", NULL );
 246 #elif defined (__FreeBSD__) || defined(__DragonFly__)
 247     static int pathname[] = { CTL_KERN, KERN_PROC, KERN_PROC_PATHNAME, -1 };
 248     size_t dir_size = PATH_MAX;
 249     dir = malloc( dir_size );
 250     if (dir)
 251     {
 252         if (sysctl( pathname, ARRAY_SIZE( pathname ), dir, &dir_size, NULL, 0 ))
 253         {
 254             free( dir );
 255             dir = NULL;
 256         }
 257     }
 258 #else
 259     dir = realpath( server_argv0, NULL );
 260 #endif
 261     if (!dir) return NULL;
 262     if (!(p = strrchr( dir, '/' )))
 263     {
 264         free( dir );
 265         return NULL;
 266     }
 267     *(++p) = 0;
 268     if (p > dir + 8 && !strcmp( p - 8, "/server/" )) nlsdir = "../nls";  /* inside build tree */
 269     if ((ret = malloc( strlen(dir) + strlen( nlsdir ) + 1 )))
 270     {
 271         strcpy( ret, dir );
 272         strcat( ret, nlsdir );
 273     }
 274     free( dir );
 275     return ret;
 276 }
 277
 278 /* load the case mapping table */
 279 struct fd *load_intl_file(void)
 280 {
 281     static const char *nls_dirs[] = { NULL, NLSDIR, "/usr/local/share/wine/nls", "/usr/share/wine/nls" };
 282     static const WCHAR nt_pathW[] = {'C',':','\\','w','i','n','d','o','w','s','\\',
 283         's','y','s','t','e','m','3','2','\\','l','_','i','n','t','l','.','n','l','s',0};
 284     static const struct unicode_str nt_name = { nt_pathW, sizeof(nt_pathW) };
 285     unsigned int i, offset, size;
 286     unsigned short data;
 287     char *path;
 288     struct fd *fd = NULL;
 289     int unix_fd;
 290     mode_t mode = 0600;
 291
 292     nls_dirs[0] = get_nls_dir();
 293     for (i = 0; i < ARRAY_SIZE( nls_dirs ); i++)
 294     {
 295         if (!nls_dirs[i]) continue;
 296         if (!(path = malloc( strlen(nls_dirs[i]) + sizeof("/l_intl.nls" )))) continue;
 297         strcpy( path, nls_dirs[i] );
 298         strcat( path, "/l_intl.nls" );
 299         if ((fd = open_fd( NULL, path, nt_name, O_RDONLY, &mode, FILE_READ_DATA,
 300                            FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE,
 301                            FILE_NON_DIRECTORY_FILE | FILE_SYNCHRONOUS_IO_NONALERT ))) break;
 302         free( path );
 303     }
 304     if (!fd) fatal_error( "failed to load l_intl.nls\n" );
 305     unix_fd = get_unix_fd( fd );
 306     /* read initial offset */
 307     if (pread( unix_fd, &data, sizeof(data), 0 ) != sizeof(data) || !data) goto failed;
 308     offset = data;
 309     /* read size of uppercase table */
 310     if (pread( unix_fd, &data, sizeof(data), offset * 2 ) != sizeof(data) || !data) goto failed;
 311     offset += data;
 312     /* read size of lowercase table */
 313     if (pread( unix_fd, &data, sizeof(data), offset * 2 ) != sizeof(data) || !data) goto failed;
 314     offset++;
 315     size = data - 1;
 316     /* read lowercase table */
 317     if (!(casemap = malloc( size * 2 ))) goto failed;
 318     if (pread( unix_fd, casemap, size * 2, offset * 2 ) != size * 2) goto failed;
 319     free( path );
 320     return fd;
 321
 322 failed:
 323     fatal_error( "invalid format for casemap table %s\n", path );
 324 }